Method bias

from Wikipedia, the free encyclopedia

The term common-method variance (Engl. Common-method bias ) referred to in the empirical distortion of the measurement results, arising from that respondents source for the same exogenous variable and the dependent variable are. The respondents can, for example, often draw conclusions about the underlying hypotheses from the questionnaire and adjust their response behavior accordingly. Since the use of the same method ( "Unit Method") for the collection of data systematic error variances can cause all the variables, the term is Einheitsmethoden variance (Engl. Common-method variance ) used to describe the phenomenon more accurately.

meaning

The importance of the uniform method variance is controversial. Some believe that it is often a problem and that researchers must do everything possible to minimize its impact. Others consider it to be a "modern legend" that is both an exaggeration and oversimplification of the real situation. If method bias can be excluded, this is an important step in demonstrating construct validity .

causes

The causes for the uniform method variance include: a. Distortions caused by ambiguous indicators, the context of the measurement, fluctuating moods of respondents, social desirability , desire for consistency ( consistency distortion ), assumed theories scale endpoints and indulgence distortion ( leniency bias ) called.

Ex ante approaches to improving a bias

The only reliable way to avoid uniform method variance is to use alternative sources of information for some of the constructs to be measured . If possible, a different source of information should be used for the dependent variable than for the independent variable . By definition, this excludes a uniform method variance. For example, two people from a company could always be interviewed or data from a database could be used for the (un) dependent variable. If this is not possible, at least some aspects can be taken into account when designing and implementing the questionnaire . So the order of the questions can be chosen with care. Different types of measuring scales can also be used. Different endpoints and formats of the scales are conceivable. Participants can also be made aware of the anonymity and confidentiality of the survey and the fact that there are neither “right” nor “wrong” answers and that they should answer as honestly as possible. Fact-based indicators may be less sensitive to unit method variance.

Ex-post approaches to identifying bias

There are a number of statistical methods with which a uniform method variance is to be recognized and even corrected ex post . In a large-scale study by Richardson et al. (2009), however, shows hardly any benefit for such methods, including a well-known method by Lindell and Whitney (2001), so that ex-post approaches are not recommended. Using an “ideal” marker variable, only one method by Williams et al. a benefit. Williams et al. (2010) understand themselves under a marking variable not only as an additional "variable that is not expected to be theoretically related to substantive variables in the model", but demand from this "capturing or tapping into one or more of the sources of bias that can occur." in the measurement context for given substantive variables being examined, given a model of the survey response process ". The authors present a three-step process with which the extent of the unit method variance can be determined using a marker variable. This method starts with the weaknesses of the Lindell and Whitney (2001) method. Williams et al. also represent requirements for the marking variable.

swell

  1. ↑ Podsakoff, PM, & Organ, DW (1986). Self-reports in organizational research: Problems and prospects. Journal of Management, 12, 69-82.
  2. ↑ Podsakoff, PM, MacKenzie, SB, Lee, JY & Podsakoff, NP (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88 (5), 879-903.
  3. ↑ Spector PE (2006): Method variance in organizational research: Truth or urban legend? Organizational Research Methods, Vol. 9, p. 230
  4. ↑ Williams, LJ, Hartman, N. & Cavazotte, F. (2010): Method Variance and Marker Variables: A Review and Comprehensive CFA Marker Technique. Organizational Research Methods, Vol. 13, No. 3, pp. 477-514.
  5. ↑ Chang et al. (2010): Common method variance in international business research. Journal of International Business Studies, Vol. 41, No. 2, pp. 178-184.
  6. ↑ Chang et al. (2010): Common method variance in international business research. Journal of International Business Studies, Vol. 41, No. 2, pp. 178-184.
  7. ↑ Richardson, HA; Simmering, MJ; Sturman, MC (2009): A tale of three perspectives: Examining post hoc statistical techniques for detection and correction of common method variance. Organizational Research Methods, Vol. 12, pp. 762-800
  8. ↑ Williams, LJ, Hartman, N. & Cavazotte, F. (2010): Method Variance and Marker Variables: A Review and Comprehensive CFA Marker Technique. Organizational Research Methods, Vol. 13, No. 3, pp. 477-514.
  9. ↑ Williams, LJ, Hartman, N. & Cavazotte, F. (2010): Method Variance and Marker Variables: A Review and Comprehensive CFA Marker Technique. Organizational Research Methods, Vol. 13, No. 3, pp. 477-514.